Printer Friendly

AFRICAN ELEPHANTS: THE EFFECT OF PROPERTY RIGHTS AND POLITICAL STABILITY.

MICHAEL L. NIESWIADOMY [*]

African elephant populations have declined by more than 50% over the past 20 years. International outrage over the slaughter led to a worldwide ban on ivory sales beginning in 1989, despite the objections of many economists and scientists, and of several southern African countries that have established systems of property rights over elephants. Far from declining, elephant populations in many of these countries have increased to levels at or above the carrying capacity of the ecosystem. This article estimates the determinants of changes in elephant populations in 35 African countries over several time periods. The authors find that, controlling for other factors, countries with property rights systems or community wildlife programs have more rapid elephant population growth rates than do those countries that do not, Political instability and the absence of representative governments significantly lower elephant growth rates. (JEL O13, Q2)

I. INTRODUCTION

The plight of the African elephant is well documented. Hunted for its valuable ivory, the African elephant population has declined dramatically, from 1.3 million in 1979, to 600,000 in 1989, to 543,000 in 1994. Fourteen African countries have lost over 60% of their elephants during this period (see Figure 1). The abhorrence of the slaughter led to the worldwide ban on the ivory trade, as mandated by the Convention on International Trade in Endangered Species of Wild Flora and Fauna (CITES) Appendix 1 listing in October of 1989. However, there was not uniform agreement for the CITES ban. Several African countries, particularly those in southern Africa, were steadfastly opposed to the ban and refused to sign the agreement. The southern African countries favor a diametrically opposite strategy for saving the elephant--a system under which property rights over the elephants are assigned. Although there are several versions of this system, they all share a common element: local Africans receive benefit from the u se of the elephants, whether for hunting safaris, for photographic safaris, or for ivory and other elephant products.

Two groups have emerged in the African elephant debate: the ivory ban group and the property rights group. The two sides each have cited data in support of their arguments. The ivory ban group notes that poaching has plummeted in recent years. The property rights groups argue that the countries that have banned ivory trade are the ones that have failed disastrously in the management of their elephants. The countries that have established property rights (primarily in southern Africa) have experienced significant increases in their elephant population.

In response to intensive lobbying by the southern African countries, and to a growing skepticism among economists and some scientists regarding the effectiveness of the ban, [1] in June of 1997, the participating CITES countries voted to relax the CITES ban effective in 1999. Zimbabwe, Namibia, and Botswana will each be permitted to sell a limited amount of ivory to Japan. (In particular, the export quotas are 25.3 tons for Botswana, 20 tons for Zimbabwe, and 13.8 tons for Namibia; Sugal, 1997b.) This relaxation amounts to an experiment: the effect of this "downlisting" on elephant populations will surely be closely watched.

Our principal purpose here is to bypass the usual rhetoric and examine this debate empirically. We review the literature in the following section, then in Section III we describe the details of the property rights systems or community wildlife programs that have been implemented. A simple empirical model is presented in Section IV. Section V describes the data, and in Section VI we present our results. A final section puts forth some conclusions.

II. LITERATURE REVIEW

An analysis of the factors influencing the African elephant population requires a review of at least two strands of literature: property rights and political/economic stability. The property rights methodology began to develop as a method of economic analysis in the 1960s. The basic tenet of the property rights approach is that individuals respond to economic incentives that are influenced by the prevailing property rights structure (Furubotn and Pejovich, 1974). Particular attention has been devoted to studying the problems of so-called "common property resources," resources that are not exclusively controlled by a single agent. The primary problem facing common property resources is the "use it or lose it" incentive. The modern development of the so-called "common property" problem is usually credited to Gordon (1954) for his discussion of the overexploitation of fisheries, to Coase (1960), and to Demsetz (1967). However, it was Garrett Hardin's influential article in Science (1968) that made "tragedy of t he commons" a household phrase among environmental researchers.

More recently, researchers have proposed using the property rights approach to save wildlife. Simmons and Kreuter (1989), Kreuter and Simmons (1995), and Sugg (1994) note that elephant populations have risen in Botswana, Namibia, South Africa, Tanzania, Zambia, and Zimbabwe, where property rights have been established. In Botswana, for example, the elephant population has risen from 20,000 in 1981 to over 80,000 today. In Zimbabwe, the roughly 30,000 elephants that existed in 1978 have increased their numbers by a factor of nearly 3 by this year. Information regarding the changes in the African elephant population is presented in the Appendix. These property rights programs are generally known as community-based natural resource management (CBNRM) programs, as described in Section III.

It should be noted that the term "common property" is often used incorrectly. More recently, property rights scholars note that there are at least three categories of property rights, not the simple two-part dichotomy of private versus common property. However, the phrase "common property" is still used often when the more correct term "open access" or "res nullis" should be used (Tietenberg, 1996). In contrast, the term "common property" implies that there are rules and regulations regarding the rights of access and usage rates of the "commons." This distinction is significant because some of the programs existing in Africa today have given tribes property rights to wildlife that could be correctly called "common property."

Many environmental groups are opposed to the property rights approach (for hunting), arguing that the ban is working (Pagel and Mace, 1991; Douglas-Hamilton, 1992). They are opposed to the hunting of elephants but are now more amenable to sharing with locals the benefit of tourism, which is a limited form of a property right.

The second strand of literature that must be examined is the political and economic climate of the African countries. Animals and humans compete for scarce resources such as land. Individual species will be preserved only if humans are willing to invest in habitat protection and services (Swanson, 1993). The economics of species extinction was first modeled by Clark (1973), who extended the work of Gordon's (1954) fishery overexploitation analysis. Clark (1973) noted that open access regimes might not yield viable stock levels. Swanson (1994) extended Clark's model to consider the choice theoretic problem of a society that wishes to maximize the value of a resource subject to the cost of allocating base resources (land) and management services to the protection of an individual species. This model yields an interesting insight into species survival: "A species must not only be capable of generating a competitive return on its own stock values, but it must also be able to earn a competitive return on the anci llary resources that it requires for its sustenance" (Swanson, 1994, p. 813).

Swanson's model can be used to explain the significant decline in the African elephant population. For elephants to survive, the social benefits of the resource need to be appropriable by members of society. However, it has been illegal to take ivory in most (but not all) sub-Saharan African countries for the past several decades. Thus, it is impossible for individual citizens to legally receive any of the benefits of the species. In essence, elephants have been treated as open-access resources. Why have some African countries banned the taking of ivory? Superficially, it would seem that African countries' leaders are trying to save the elephant by banning the ivory trade. However, a more likely explanation is that these leaders need favorable international public opinion to continue to trade with and receive aid from the industrial world. Banning the ivory trade helps bolster the image of their respective countries. Yet at the same time, these leaders often tacitly allow the harvesting of ivory, then receiv e revenue from the seizure of illegal ivory (Swanson, 1989).

But Swanson's (1994) model also has an additional interpretation. Many of these African leaders have seized power ruthlessly. Their leadership positions have been tenuous and are susceptible to a sudden loss of power. In some cases they have fled their countries, taking their wealth with them. With their future reigns so uncertain, their discount rate is quite high. The present value of any future flows of ivory revenues is likely to be relatively small. A leader is likely to allow ivory to be taken (and receive his share) as quickly as possible. He does not necessarily have the long-term best interests of the country at heart. [2] Even if citizens had some limited property rights to the elephants, their discount rate is also likely to be high because of the instability in their governments.

Two recent studies highlight the importance of stable governments for protecting elephants and rhinos. Kremer and Morcom (1996) note that stable, credible governments are important in protecting species. They show that "if governments have credibility, they may be able to eliminate the extinction equilibrium" by "promising to implement tough anti-poaching measures if the population falls below a threshold" (pp. 4--5). An alternative strategy involves "building sufficient stockpiles of the storable good, and threatening to sell the stockpile if the animal becomes endangered or the price rises beyond a threshold" (p. 5). For this strategy to be successful, the threats must come from credible and stable governments. Brown and Layton (1998) devised an interesting model for saving the rhino by supplying horns without killing the rhinos. The horns are cut off and allowed to regrow. The price of the horn is controlled below the opportunity cost of poachers. However, this strategy is only effective if the government s are stable enough to maintain this policy consistently. The three countries that they recommend for implementing this policy are the only ones in Africa with any significant rhino populations (South Africa, Namibia, and Zimbabwe). These three countries have been in the vanguard of establishing property rights for wildlife, as discussed below.

In a similar vein, Deacon (1994) has shown that deforestation has occurred more rapidly in countries facing political unrest (e.g., political assassinations, riots, coups d'etats, etc.) or that have nonrepresentative governments (e.g., military dictator). He surmises that these factors tend to reduce the security of property rights, thereby causing individuals to focus on the short-run benefits while ignoring long-run consequences. However, no one has examined systematically the impact of the economic and political environment on property rights and the attendant effects on the African elephant.

III. COMMUNITY-BASED NATURAL RESOURCE MANAGEMENT PROGRAMS

CBNRM programs of one sort or another now exist in at least eight African countries. The CBNRM idea has become increasingly popular because of the general failure of the state-controlled national parks model. In that model, areas were set aside as preserves controlled by the central government. Residents of the area were abruptly forbidden to utilize the resources of their surroundings, despite the fact that such utilization had occurred for many generations. Any benefit from the existence of wildlife that once existed for local residents now accrued to the state. These losses were in many cases substantial: not only could local residents not profit from the sale of animal products (meat, hides, tusks, etc.), but also they lost income that they might have earned as guides to safari hunters. Furthermore, local residents lost the ability to harvest animals for purposes of subsistence consumption (Barbier, 1992).

While local residents lost much of the use value of the wildlife, their homes and crops, and occasionally their lives, were threatened by wildlife. It should come as no surprise that subsistence farmers commonly view elephants as pests: they receive little or no benefit from and yet must bear the costs of their existence (Bonner, 1993; Sugal, 1997a). Indeed, given the tremendous poverty of rural Africa, it is easy to see how local residents might be tempted to harvest elephants and other wildlife illegally, or to assist poachers. Certainly, the residents would have no reason to impede the activities of poachers, who might even be seen as providing a much-needed service to the local populace.

The CBNRM movement has attempted to return some of the existence benefits of elephants and other wildlife to the local people. According to Barbier (1992), CBNRM programs are normally characterized by a combination of several sorts of benefits. First, revenue from such activities as tourism, safari hunting, and the sales of wildlife-related products is typically shared with the surrounding community. Second, jobs and income are frequently generated for the community. Third, some of the revenue generated by CBNRM funds community improvements such as schools, water pumps, and roads. Finally, CBNRM often allows local residents limited rights to use the wildlife themselves, most typically for meat and hides. Now that the CITES ban on ivory sales has been partially relaxed, villages may hope to raise revenues from the sale of ivory as well.

The details of particular CBNRM programs vary. Three of the countries, Namibia, Zimbabwe, and South Africa, have codified property rights to wildlife in national law. Other countries (Zambia, Tanzania, Kenya, and Botswana) have established localized community wildlife/aid programs that vary in type and quality. These programs may be less effective than the national law programs, but they all share one or more of the above characteristics. A description of these programs is available from the authors on request.

IV. THE MODEL

Our approach is primarily an empirical rather than a theoretical one. The variable we seek to explain is the population of elephants over time and across countries. Using an approach similar to that of Deacon (1994), we hypothesize that the number of elephants, E, depends on several general categories of factors. First, economic characteristics, Y, may affect the number of persons interested in poaching. Y may include per capita income, since poorer people may be more inclined to seek poaching rents. Y may also include the price of ivory. The dramatic increase in the ivory price in the late 1980s was coincident with a tremendous decline in elephant populations (Caldwell, 1998; Caldwell and Luxmoore, 1990). Second, anecdotal evidence suggests that political conditions, P, influence a given country's ability to control rampant poaching. For example, political instability in Uganda is thought to have led to wholesale slaughter of elephants in Murchison Falls National Park as soldiers sought ivory to finance the ir activities (Douglas-Hamilton, 1992). It is also likely that the level of corruption of government officials affects elephant populations. Given the incredibly lucrative nature of the ivory trade, government officials charged with controlling poaching may be tempted to accept bribes to "look the other way." Third, weather conditions, such as the frequency and duration of droughts, D, surely affect wildlife populations over time and across countries. Fourth, we postulate that the success of antipoaching programs (and thus the populations of elephants) depends importantly on the degree to which local peoples have ownership on their lands or community wildlife laws, O. As mentioned above, there are two types of CBNRM programs. [3] A casual perusal of the data contained in Appendix Table 1 confirms that in many of these countries elephant populations are growing rapidly. Finally, the percentage of a country devoted to protection of wildlife, W, is likely to be an important variable (Swanson, 1994). The function can be expressed as follows:

(1) [E.sub.it] = f([Y.sub.it], [P.sub.it], [D.sub.it], [O.sub.it], [W.sub.it], [[beta].sub.it]),

where i and t index countries and years, respectively, and [beta] is a parameter vector.

Modeling the level of elephant populations across countries is very difficult because many historical factors determine how a population reaches a certain level. A pragmatic approach is to analyze the percentage change in the elephant population during various intervals over the past 25 years. Taking the logs of Equation (1) and first differencing between two time periods yields

(2) [D.sub.i] = log[f([Z.sub.i,t];[[beta].sub.i,t])]

- log[f([Z.sub.i,t-1];[[beta].sub.i,t-1])],

where [D.sub.i] = [log([E.sub.i,t]) -- log ([E.sub.i,t-1])] represents the proportional change in the elephant population. We further assume that coefficients are the same across time periods and that log(*) from Equation (2) is a linear function of the explanatory variables. [4]

V. THE DATA

Systematic estimates of elephant populations over time in each African country (with a significant number of elephants) have been produced only in recent years. Iain Douglas-Hamilton is the preeminent authority on the African elephant. He initiated the first African elephant counts in 1976 and has made periodic updates. These estimates usually are based on either aerial surveys or dung counts. In particular, we use Douglas-Hamilton's population figures for the 1976-1979 and 1989 periods (Elephants, 1979; Douglas-Hamilton et al., 1992). The 1994 population estimates are taken from the 1995 African Elephant Database (Said et al., 1995), which built on the work of Douglas-Hamilton. The 1981 figures are taken from the African Elephant Action Plan (1990). Data from earlier periods were used if the source was deemed reliable enough. For example, Douglas-Hamilton's (1992) estimates of Uganda's elephant populations in 1969 and 1973 are included in our analysis. The population estimates used in this research, as well as additional information on data sources, can be found in the Appendix.

We control for the CITES ban by including a dummy variable taking on the value of one for periods since 1989 and zero otherwise. Changes in climatological conditions are modeled using the U.S. National Weather Service's Climate Anomaly Monitoring System (CAMS) database, which includes temperature and precipitation data from 500 African stations over the entire period of interest. We use a crude drought index that measures the percentage of time that a country had monthly rainfall 50% or more below the historical monthly averages.

One of our hypotheses is that countries that have created systems by which local peoples have some degree of proprietorship over the wildlife on their lands will more easily control poaching. Because of the significant differences in wildlife programs, we model property rights using two dummy variables. The first, PROPERTY, takes on the value of one for every country that had a national-level natural resource management program that assigned some property rights to local communities during the time period in question. In particular, this variable takes on a value of one for Namibia during the 1981-89 and 1989-94 periods and for South Africa and Zimbabwe for 1978-81, 1981-89, and 1989-94 periods. The second dummy variable, COMMAID, controls for the several countries that do not have national programs establishing local control over wildlife, but instead have localized CBNRM programs. This variable takes on the value of one for Botswana during the 1981-89 and 1989-94 periods, and for Kenya, Tanzania, and Zambi a for the 1989-94 period.

As discussed above, political instability and nonrepresentative forms of government may have independent effects on elephant populations. Using the Cross-National Time Series Archive, 1997 (Banks, 1997) we constructed a number of measures of these factors. A citizen's perception of the security of property rights in a given year is influenced not only by the political stability in that year, but also by past stability. If, for example, a country had experienced political instability two years ago, citizens may still feel insecure about property rights. Thus, we calculated the average level of these variables for the time period corresponding to an elephant population measure and five years of lagged values of the political variables. [5] Measures of political instability include the average yearly number of assassinations, coups, antigovernment demonstrations, constitutional changes, riots, major government crises, strikes, and regime changes in the current period plus the preceding five years. Some of the m easures of nonrepresentativeness of government include whether or not the executive is military, whether or not the executive is elected, and the number of political purges in the most recent period. (For the indicator variables, such as the military executive indicator, we calculated the percentage of time that the indicator was equal to one for the time period including five lagged years. For the other nonrepresentativeness variables, we calculated the average annual number of events for the time period plus five lagged years.) The mean of each of these variables is presented in Table 1 for countries with shrinking and countries with growing elephant populations.

The price of ivory obviously is an important variable impacting the population of elephants. However, the price of ivory is impossible to determine after the 1989 CITES ban was implemented. Futhermore, the price of ivory received by poachers varies considerably across countries (Barbier et al., 1992). There are not reliable data to accurately measure this variable.

Finally, it may be that countries with greater amounts of protected area have faster growing elephant populations. Using data from the World Conservation Monitoring Centre (WCMC), we construct a variable (PROAREA) that is the average percentage of each country's land area that is protected over each time period. For example, between 1981 and 1989 an average of 5.45% of Kenya's land area was categorized by the International Union for the Conservation of Nature and Natural Resources (IUCN) as protected; between 1989 and 1994, this rose to 5.77%. [6]

VI. THE RESULTS

Because the political variables are quite collinear, we first examine some simple comparison of means for countries that had increasing elephant populations versus those that had decreasing populations. Since the political climate can change in a country over time, each time period for each country was treated as a different observation. Some countries had decreasing populations for part of the time but increasing populations for other time periods. This dichotomy of positive and negative growth rates yields 52 observations with declining populations and 23 with increasing populations.

The tests of difference of means are shown in Table 1. Most of the results are as expected. Three of the political instability measures are significant at the 10% level. Countries with declining elephant populations have a higher number of strikes, government crises, and regime changes. Specifically, countries with shrinking elephant populations have three times as many strikes, twice as many government crises, and nearly twice as many regime changes.

Only one of the measures of government nonrepresentativeness is statistically significant at the 10% level. Countries with declining elephant populations have more purges (specifically, three times as many), as is expected. These comparisons of mean attributes of political instability and of government nonrepresentativeness indicate that elephants are more threatened when the security of property rights is attenuated.

To capture the effects of several factors that may affect elephant population growth rates, we use a basic linear regression model. (In addition, data on international reserves were used as a measure of wealth but the coefficient was insignificant. Several countries did not have data on this variable.) Since the political variables are collinear and we have a limited number of observations, we use only one measure of political instability (riots) and one measure of nonrepresentativeness (purges). Many other political instability variables were also statistically significant, but the number of purges was the only measure of nonrepresentativeness that was significant. We estimated two versions of the model. One model includes both the dummy variable representing whether or not a country has a national-level property rights system and the dummy variable representing the more localized community-based programs. It can be argued that countries with localized wildlife aid programs vary substantially in quality and coverage. Categorizing these countries with one dummy variable (COMMAID) may be problematic. Therefore, the other model includes only the national property rights dummy (PROPERTY).

The results of the first model are shown in Table 2. In general, the coefficients have the expected signs. Countries that had more riots or purges during a given period experienced a decrease in their elephant population, although the coefficient on riots is not statistically significant. The coefficient on the number of purges indicates that an increase of one purge per year decreases the elephant population growth rate by 14.1 percentage points. The fact that a purge has a much stronger effect than a riot seems reasonable since a purge generally represents more instability than a riot.

The national-level property rights dummy variable also has the expected sign and is highly statistically significant. Countries that have national programs that vest some ownership of wildlife in their citizens have experienced more than a 19% higher annual growth rate than countries that do not. The dummy variable for the localized community property rights programs is also positive and highly significant. Holding other factors constant, countries with these local community wildlife programs have populations that grow nearly 17% faster than those that do not. Indeed, these local programs seem to be nearly as effective as the national law programs in South Africa, Namibia, and Zimbabwe. Effects of these magnitudes may seem quite large, but it should be noted that many of the growth rates are negative and large (in absolute value) for those countries that have declining populations. On the other hand, the growth rates are positive (or close to zero) for those countries that have increasing populations.

The CITES ban dummy variable was not significant in any of the models estimated. This does not mean that the ban had no effect; rather, the effects of the other variables may have been so strong that any impact of the CITES ban may have been difficult to ascertain. In fact, several countries have continued to experience measured declines in elephant populations during the 1990s due to political turmoil. There are also some countries that are believed to have experienced continued declines in the 1990s, but observers have not been able to enter the countries (without extreme risks) to measure the populations. [7]

The drought variable was also not significant. As with the CITES ban variable, this insignificance does not imply that climatological factors have no influence on elephant populations. It may be simply that the impacts of the political and property fights variables were overwhelmingly strong. Obviously, poachers could slaughter elephants at enormously high rates whereas deaths due to unfavorable climatic conditions would occur at a substantially slower rate.

Although the variable measuring the percentage of land protected does have the expected positive sign, it is not significantly different from zero.

In Table 3, similar results are presented. This model was estimated using only the national-level property rights dummy. The same basic pattern of results holds for this model. In particular, the effect of property rights is positive and significant. The effect of purges is negative and significant. Drought and the percentage of land protected continue to be statistically insignificant variables. This model demonstrates the robustness of the effects of the property rights and political variables.

VII. CONCLUSIONS

The fate of the African elephant may be determined within the next decade. Many of the older elephants (particularly males) have been killed, leaving many herds without leadership. Concern for the fate of the elephants has led to at least two different strategies. One strategy is to ban all sale of ivory. The other suggests deeding some property rights to the local Africans. Local Africans bear the burden of the elephants that sometimes destroy crops and damage property. Adjudicating to them property rights to the elephants for tourism and sale of elephant products (tusks and hides) gives them incentives to protect the elephants. [8]

This article has examined the impact of several factors on elephant population growth rates. Comparisons of means of measures of political instability and government nonrepresentativeness indicate that political instability is partially responsible for declining elephant populations. Countries with declining elephant populations have more strikes, major government crises, and regime changes, as well as more purges. A regression analysis indicates that purges negatively impact elephant population growth rates. Both of the property rights dummy variables were highly significant, indicating that elephant populations are more likely to thrive in countries that vest in their citizens some proprietary rights to the elephants. The CITES ban does not have any statistically significant effect. However, the ban may have provided some positive impact for the non-southern African countries that is difficult to quantify. Neither our drought index nor our measure of protected areas was significant, perhaps due to the over whelming effects of the property rights and political variables.

There are several caveats that must be given. Since measuring elephant populations is difficult, the magnitudes of the estimated coefficients are not precise. However, the negative impact of the political instability variables seems quite robust, even if the exact magnitude may be difficult to pinpoint. The impact of the property rights is quite compelling. Nonetheless, establishing property rights in other African countries is not an easy task. Several of the southern African countries have demonstrated that they can sustainably use their elephants. They can use the proceeds from sale of ivory to better protect their elephants. In fact, the continued protection of the elephant depends on this source of revenue (Barnes, 1996; Khanna and Harford, 1996). Foreign donors have not helped these countries during the CITES ban, even though they benefit from the existence value of the elephant. The limited relaxation of the CITES ban agreed on in June 1997 for Botswana, Namibia, and Zimbabwe should benefit these nati ons.

Increasing political stability and representative government must accompany the establishment of property rights. However, many African countries are so rife with turmoil that it is not feasible to establish property rights at this time. Furthermore, since the range of the elephants is substantial, it is not possible to totally protect them from threats in unstable surrounding countries. The only viable strategy to save the elephant in these regions is to foster greater political stability and more representative government in these countries. This remains a daunting task in some countries.

McPherson: Associate Professor of Economics, Economics Department, University of North Texas, Denton, Phone 940-565-2270, Fax 940-565-4426, Email mcpherson@econ.unt.edu

Nieswiadomy: Professor of Economics, Economics Department, University of North Texas, Denton, Phone 940-565-2244, Fax 940-565-4426, E-mail mike@econ.unt.edu

(*.) This is a revision of a paper presented at the 73rd annual Western Economic Association International Conference, Lake Tahoe, Nevada, July 2, 1998, in a session chaired by Margriet F. Caswell, USDA Economic Research Service. The authors gratefully acknowledge helpful comments from three referees, as well as advice and assistance from Timothy M. Swanson, John R. Caldwell, Randy T. Simmons, and lain Douglas-Hamilton. The authors are also grateful for the assistance and cooperation of the staff of the U.S. National Weather Service's National Centers for Environmental Prediction, who provided the climate data used in this article.

(1.) Kaempfer and Lowenberg (1998) argue that the CITES ban amounts to an antitrade policy imposed by countries with declining elephant populations on countries with sound wildlife management programs. They go on to say that "attempting to abolish this value [of ivory] by denying the availability of ivory to the marketplace is futile because a market will always arise for any valued good, whether that market is legal or not" (p. 18).

(2.) In a related problem, it has been argued that U.S. oil companies pumped oil out of Middle East countries at rates faster than that in the best interest of these countries because the oil companies realized their position was tenuous (Griffin, 1985).

(3.) CBNRM programs may themselves be determined by political and other factors, leading to possible endogeneity problems. Unfortunately, data limitations preclude any effort to control for this problem.

(4.) In a similar circumstance, Deacon (1994) modeled deforestation rates, rather than levels of forest biomass, and assumed that log(*) was a linear function of the explanatory variables. Furthermore, since many of our variables have zero values, taking logs was not possible.

(5.) Deacon (1994) used separate variables for contemporaneous and lagged values of the political variables. His data all came from the same five-year time period. Our data cover time periods of different lengths, making it difficult to use a simple rule for lagging variables. We also have a limited number of observations on elephant populations.

(6.) It could be argued that central government expenditure on protected areas is a determinant of elephant population growth. However, data are only available for a relatively small number of African countries in the 1980s (James et al., 1997). As an alternative measure, we use percentage of the surface area that is protected.

(7.) It is also possible to examine the effect of the CITES ban using a Chow test for structural change. The null hypothesis of structural stability between the preban and postban periods was tested for the regressions presented in Tables 2 and 3. The F-statistics for these Chow tests are 0.509 and 0.491, so the null hypothesis of structural stability cannot be rejected at any reasonable error level.

(8.) There is some concern that elephants will be sold for ivory rather than maintained for tourism if the CITES ban is lifted, as a result of higher ivory prices. However, Swanson (1989) notes that elephant populations may be maintained for tourism or sustainable ivory harvest if they yield a competitive return on the management assets. The price inelastic demand of ivory facilitates a competitive return. In fact, the sustainable use of ivory appears to have occurred in several of the southern African countries before the ban. Also, Kremer and Morcom (1996) show that it is possible to eliminate extinction equilibria if the government promises to implement tough antipoaching measures or if private agents accumulate a sufficient stockpile of the storable good (ivory).

REFERENCES

African Elephant Action Plan, prepared under the auspices of the African Elephant Conservation Coordinating Group, 4th edition, Nairobi, March 1990.

Banks, A. S., Cross-National Time Series Data Archive, 1997, Binghamton, N.Y., 1997.

Barbier, Edward B., "Community-Based Development in Africa," in Economics for the Wilds, Timothy M. Swanson and Edward B. Barbier, eds., Island Press, Washington, D.C., 1992.

Barbier, Edward B., Joanne C. Burgess, Timothy M. Swanson, and David W. Pearce, Elephants, Economics and Ivory, Earthscan Publications, London, 1990.

Barnes, J. I., "Changes in the Economic Use Value of Elephant in Botswana: The Effect of International Trade Prohibition," Ecological Economics, 18, 1996, 215-230.

Bonner, Raymond, At the Hand of Man: Peril and Hope for Africa's Wildlife, Knopf, New York, 1993.

Brown, Gardner, and David Layton, "Saving Rhinos," Working paper, Department of Economics, University of Washington, Seattle, 1998.

Caldwell, J. R., CITES Trade Database Manager, World Conservation Monitoring Centre, Cambridge, personal communication, 1998.

Caldwell, J. R., and R. A. Luxmoore, "Recent Changes in World Ivory Trade," Traffic Bulletin, 11:2/3, March 1990, 50-58.

Clark, Colin W., "Profit Maximization and the Extinction of Animal Species," Journal of Political Economy, 81, 1973, 950-961.

Coase, Ronald H., "The Problem of Social Cost," Journal of Law and Economics, 3, 1960, 144.

Deacon, Robert T., "Deforestation and the Rule of Law in a Cross-Section of Countries," Land Economics, 70:4, November 1994, 414-430.

Demsetz, Harold, "Toward A Theory of Property Rights," American Economic Review, 57, May 1967, 347-359.

Douglas-Hamilton, Iain, Battle for the Elephants, Viking Penguin, New York, 1992.

Douglas-Hamilton, Iain, Fran Michelmore, and Amar Inamdar, African Elephant Database, United Nations Environment Programme, New York, February 1992.

Elephants: Hearings Before the Committee on Merchant Marine and Fisheries, House of Representatives, Ninety-Sixth Congress First Session on H.R. 4685, U.S. Government Printing Office, Washington, D.C., 1979.

Furubotn, Eric, and Svetozar Pejovich, The Economics of Property Rights, Ballinger Publishing, Cambridge, Mass., 1974.

Gordon, H. Scott, "The Economic Theory of a Common Property Resources: The Fishery," Journal of Political Economy, 62, 1954, 124.

Griffin, James M., "OPEC Behavior: A Test of Alternative Hypotheses," The American Economic Review, 75, December 1985, 954-963.

Hardin, Garrett, "The Tragedy of the Commons," Science, 162, 1968, 1,243-1,248.

James, Alexander N., Michael J. B. Green, and James R. Paine, "Financial Indicators and Targets for Protected Areas," Working paper, World Conservation Monitoring Centre, Cambridge, 1997.

Kaempfer, William H., and Anton D. Lowenberg, "The Ivory Bandwagon: A Case Study in the International Transmission of Interest-Group Politics," draft paper, University of Colorado, Boulder, July 1998.

Khanna, Jyot, and Hon Harford, "The Ivory Trade Ban: Is It Effective?" Ecological Economics, 19, 1996, 147-155.

Kremer, Michael, and Charles Morcom, "Elephants," NBER Working Paper 5674, Cambridge, Mass., 1996.

Kreuter, Urs P., and Randy T. Simmons, "Who Owns the Elephants? The Political Economy of Saving the African Elephant," in Wildlife in the Marketplace, Terry L. Anderson, and Peter J. Hill, eds., Rowman and Littlefield Publishers, Lanham, Md., 1995.

Pagel, Mark, and Ruth Mace, "Keeping the Ivory Trade Banned," Nature, 351:23, May 1991, 265-266.

Said, M. Y., R. N. Chunge, G. C. Craig, C. R. Thouless, R. F. W. Barnes, and H. T. Dublin, African Elephant Database, 1995, United Nations Environment Programme, New York, 1995.

Simmons, Randy T., and Urs P. Kreuter, "Herd Mentality: Banning Ivory Sales Is No Way to Save the Elephant," Policy Review, 50, Fall 1989, 46-49.

Sugal, Cheri, "Can 'CAMPFIRE' Save the Elephants?" A Woridwatch Report, The Worldwatch Institute, Washington, D.C., 1997a.

_____, "Elephants of Southern Africa Must Now 'Pay Their Way'," A Worldwatch Report, The Worldwatch Institute, Washington, D.C., 1997b.

Sugg, Ike C., "Appropriate Authority," Community Enterprise Institute Update, 7:3, March 1994, 1-3.

Swanson, Timothy M., "The International Regulation of the Ivory Trade," LEEC Paper 89-04, International Institute for Environment and Development/University College London Environmental Economics Centre, May 1989.

_____, "Endangered Species," Economic Policy, 8:16, April 1993, 183-205.

_____, "The Economics of Extinction Revisited and Revised: A Generalised Framework for the Analysis of the Problems of Endangered Species and Biodiversity Losses," Oxford Economic Papers, 46:5 (suppl.), October 1994, 800-821.

Tietenberg, Tom, Environmental and Natural Resource Economics, HarperCollins Publishers, New York, 1996.
 Changes in Elephant Population Growth Rates versus
 Political Instability and Government
 Nonrepresentativeness (mean indicators by elephant growth rate)
 Countries with Countries with
 Shrinking Growing
 Elephant Elephant
 Populations Populations
Measures of Political Instability
 Political assassinations .0642 .0784
 Strikes .0395 .0148
 Riots .1935 .1787
 Antigovernment .1433 .1325
 demonstrations
 Guerilla warfare .1524 .1849
 Revolutions .2734 .2448
 Major government crises .0931 .0537
 Coups d'etat .0547 .0466
 Constitutional changes .1406 .1385
 Regime changes .0669 .0404
Measures of Government Nonrepresentativeness
 Executive is military .1458 .0968
 Executive is not elected .4380 .3473
 No Legislature exists .3320 .2623
 Executive is not a premier .5626 .5548
 Purges .0944 .0312
 Changes in executive .1001 .1292
Number of observations 52 23
 t-Statistic
Measures of Political Instability
 Political assassinations -0.42
 Strikes 1.49
 Riots 0.21
 Antigovernment 0.28
 demonstrations
 Guerilla warfare -0.45
 Revolutions 0.45
 Major government crises 1.37
 Coups d'etat 0.40
 Constitutional changes 0.07
 Regime changes 1.51
Measures of Government Nonrepresentativeness
 Executive is military 0.76
 Executive is not elected 0.97
 No tegislature exists 0.78
 Executive is not a premier 0.08
 Purges 1.86
 Changes in executive -0.95
Number of observations
 p-Value
 (Pr [greater than] t)
Measures of Political Instability
 Political assassinations .3385
 Strikes .0705
 Riots .4155
 Antigovernment .3920
 demonstrations
 Guerilla warfare .3285
 Revolutions .3290
 Major government crises .0885
 Coups d'etat .3465
 Constitutional changes .4735
 Regime changes .0685
Measures of Government Nonrepresentativeness
 Executive is military .2265
 Executive is not elected .1700
 No tegislature exists .2210
 Executive is not a premier .4690
 Purges .0340
 Changes in executive .1760
Number of observations --


Notes: p-Values refer to the one-tailed test of the hypotheses that for each variable the mean for countries with growing elephant populations is at least as large as the mean for the shrinking population countries.

Data include 34 African countries. (Most countries have elephant population changes for 1981-89 and 1989-94. A few countries have data for some earlier periods.) Given the tremendous amount of unrest in South Africa during this period of time, observations from that country are excluded to avoid skewing the data.

Source: Banks, Arthur S., Cross-National Time Series Data Archive, 1997.
 Elephant Population Growth Rates: The Effect of the
 CITES Ban, Riots, Purges, Droughts,
 Property Rights, Community Aid Programs, and Protected Areas
Variable Coefficient Standard Error t-Ratio [Pr [greater than] t] Mean of X
Constant -0.1191 0.0727 -1.636 0.11
CITESBAN -0.0473 0.0390 -1.212 0.23 0.4231
RIOTS -0.0107 0.0086 -1.254 0.21 0.4335
PURGES -0.1409 0.0400 -3.524 0.00 0.0771
DROUGHT 0.1081 0.1915 0.565 0.57 0.3011
PROPERTY 0.1929 0.0463 4.163 0.00 0.1026
COMMAID 0.1699 0.0408 4.165 0.00 0.0641
PROAREA 0.0026 0.0050 0.513 0.61 6.3620


Dependent variable is ELEGROW

Model size: observations = 78

Residuals: sum of squares = 1.7538

Fit: [R.sup.2] = 0.2101

Model test: F(6, 71) = 2.66

Results corrected for heteroskedasticity

Mean = -0.075

Parameters = 8

Std. Dev. = 0.1583

Adjusted [R.sup.2] = 0.1311

Prob value = 0.0169

Std. Dev. = 0.1698

d.f. = 70
 Elephant Population Growth Rates: The Effect of the
 CITES Ban, Riots, Purges, Droughts,
 Property Rights, and Protected Areas
Variable Coefficient Standard Error t-Ratio [Pr [greater than] t] Mean of X
Constant -0.1712 0.0693 -2.469 0.02
CITESBAN -0.0335 0.0399 -0.838 0.40 0.4231
RIOTS -0.0056 0.0080 -0.702 0.49 0.4335
PURGES -0.1562 0.0383 -4.083 0.00 0.0771
DROUGHT 0.2094 0.1826 1.147 0.26 0.3011
PROPERTY 0.1521 0.0436 3.492 0.00 0.1026
PROAREA 0.0073 0.0048 1.520 0.13 6.3620


Dependent variable is ELEGROW

Model size: observations = 78

Residuals: sum of squares = 1.851

Fit: [R.sup.2] = 0.1662

Model test: F(5, 39) = 2.36

Results corrected for heteroskedasticity

Mean = -0.075

Parameters = 7

Std. Dev. = 0.1615

Adjusted [R.sup.2] = 0.0957

Prob value = 0.0392

Std. Dev. = 0.1698 d.f.= 71

ABBREVIATIONS

CAMS: Climate Anomaly Monitoring System

CBNRM: Community-based natural resource management

CITES: Convention on International Trade in Endangered Species of Wild Flora and Fauna

IUCN: International Union for the Conservation of Nature and Natural Resources

WCMC: World Conservation Monitoring Centre

APPENDIX

Elephant Populations and Sources

It is notoriously difficult to accurately measure elephant populations. This is due to many factors, including the fact that elephants roam over large and remote areas. In practice, elephants are counted in different ways according to the country involved and the particular time period. For example, aerial counting methods are often used in savanna areas, whereas the somewhat less reliable dung counts are more commonly used in densely forested areas. We recognize the variability in the quality of data across both countries and time; our dependent variable is our best effort to collect data from various sources. This Appendix is an effort to describe the elephant population data in greater detail, so that the reader can make independent decisions regarding the quality of the data.

The elephant population numbers used to calculate the dependent variable in our regression are presented in Appendix Table 1. The sources of the data and a brief discussion of the details of each data source are presented below.

1994: These figures come from the African Elephant Database, 1995 (see Said et al., 1995), which was generated through collaborative effort between the World Conservation Union and the United Nations Environment Programme. The authors separate the quality of the population estimate for each country into four categories according to how the populations were measured: definite, probable, possible and speculative. In general, total counts of elephants conducted either on the ground or aerially are classified in the "definite category"; however, sample counts (aerial or ground), as the primary source of the estimate, are classified in the "probable" category. Should the estimates rely heavily on the less accurate dung count method, a larger proportion of the estimate will fall in the "possible" category. Least reliable of all are the "speculative" estimates, which are based on informed guesses. The interested reader should refer to Said et al. (1995, pp. 10-11) for further details. Where possible, we use the sum of the "definite," "probable," and "possible" categories. If only the "speculative" figure is available, we used that. The countries for which only speculative data were available are as follows: Angola, Guinea, Liberia, and Niger. The countries with the highest ratio of "definite" and "probable" figures to the total population are Botswana, Burkina Faso, Gabon, Kenya, Malawi, Mali, Namibia, South Africa, Tanzania, Uganda, Zambia, and Zimbabwe. These figures can therefore be regarded as the most reliable.

1989: The 1989 figures come from lain Douglas-Hamilton's 1989 Ivory Trade Review Group report to the CITES conference, October 1989, as reported in the African Elephant Action Plan, prepared under the auspices of the African Elephant Conservation Coordinating Group, March 1990. His data come from various aerial and ground surveys, dung counts, and informed guesses. Although these data appear to have been compiled in a less systematic manner than the 1994 numbers, these figures should be regarded as the "best guess" of one of the world's preeminent authorities on African elephants.

1981: These data come from the African Elephant and Rhino Specialist Group of the IUCN, as reported in the African Elephant Action Plan (1990). Once again, although they are surely flawed, they represent the most accurate available estimates of national elephant populations assembled by the world's leading elephant experts.

1978: Once again, these data are Iain Douglas-Hamilton's estimates, assembled under the auspices of the IUCN and the World Wildlife Fund. These particular numbers were part of his presentation to the U.S. House of Representatives Committee on Merchant Marine and Fisheries in 1979.

Pre-1978: All five of the earlier population estimates were taken from lain Douglas-Hamilton's 1992 book, Battle for the Elephants.
 Elephant Populations, 1969-1994
Country 1969 1970 1973 1976 1978 1981
Angola
Benin 1,250
Botswana 20,000
Burkina Faso 3,500
Cameroon
Central African Republic 100,000 65,800 31,000
Chad 15,000
Congo
Equatorial Guinea 1,300
Ethiopia
Gabon 13,400
Ghana 2,400
Guinea 800
Ivory Coast 4,800
Kenya 130,000 65,056
Liberia 2,000
Malawi 4,500
Mali 1,000 780
Mauritania 164 40
Mozambique 54,800
Namibia 2,300
Niger 1,500 800
Nigeria 1,820
Rwanda 150
Senegal 450 200
Sierra Leone 500
Somalia 24,323
South Africa 7,800 8,000
Sudan 133,727
Tanzania 350,000 203,900
Togo 150
Uganda 40,000 20,000 5,900 2,320
Zaire 376,000
Zambia 216,280 160,000
Zimbabwe 30,000 49,000
Country 1989 1994
Angola 18,000 8,170
Benin 2,100 1,400
Botswana 68,000 80,174
Burkina Faso 4,500 2,635
Cameroon 22,000 16,613
Central African Republic 23,000 4,390
Chad 2,100 1,040
Congo 42,000 32,563
Equatorial Guinea 500 407
Ethiopia 8,000 2,407
Gabon 74,000 82,012
Ghana 2,800 2,515
Guinea 560 1,000
Ivory Coast 3,600 1,611
Kenya 16,000 25,554
Liberia 1,300 1,783
Malawi 2,800 2,087
Mali 840 762
Mauritania 100 0
Mozambique 17,000 1,495
Namibia 5,700 11,999
Niger 440 800
Nigeria 1,300 1,065
Rwanda 50 71
Senegal 140 20
Sierra Leone 380
Somalia 2,000 130
South Africa 7,800 10,010
Sudan 22,000
Tanzania 61,000 98,179
Togo 380 85
Uganda 1,600 1,848
Zaire 112,000 83,618
Zambia 32,000 33,004
Zimbabwe 52,000 81,855
COPYRIGHT 2000 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:MCPHERSON, MICHAEL A.; NIESWIADOMY, MICHAEL L.
Publication:Contemporary Economic Policy
Article Type:Statistical Data Included
Geographic Code:60AFR
Date:Jan 1, 2000
Words:7989
Previous Article:WORLD OIL: THE GROWING CASE FOR INTERNATIONAL POLICY.
Next Article:WHAT DO SCHOOLS PRODUCE? IMPLICATIONS OF MULTIPLE OUTPUTS IN EDUCATION.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters